Match fails oddly if branch is changed from int to i64 about v HOT 5 CLOSED

I agree. Auto promoting int to i64 is fine, since it an int fits entirely in the range of i64, but automatically casting i64 to int implicitly, should not be allowed.

from v.

For my own understanding:

Why doesn't V report an error when the first branch is an int and the other branches are i64?

The current behavior suggests that V is silently casting the 2nd and subsequent branches to int.

from v.

Yes. Because the first branch on Match is instructing that the return type is int not i64.

You can use: 'b' { i64(block) } to force i64 too.

assert string_to_i64('18T')! == 18 * terra

With int:
number: 18
multiplier: 0
result: 0

With i64:
number: 18
multiplier: 1099511627776
result: 19791209299968

from v.

Yes. Because the first branch on Match is instructing that the return type is int not i64.

Why does the p branch work?

Also, I would expect multiplier to always be i64. Otherwise, the other branches should be errors since they are i64

from v.

Why does the p branch work?

Ah, I defined peta wrong. bummer.

Still seems odd that match does not issue an error in the other branches beause of assignment of an i64 to int

from v.